conscious and unconscious memory...

Conscious and Unconscious Memory Systems

Larry R. Squire1,2,3 and Adam J.O. Dede3

1Veterans Affairs, San Diego Healthcare System, San Diego, La Jolla, California 921612Departments of Psychiatry and Neurosciences, University of California, San Diego, La Jolla, California 920933Department of Psychology, University of California, San Diego, La Jolla, California 92093

Correspondence: [email protected]

The idea that memory is not a single mental faculty has a long and interesting history butbecame a topic of experimental and biologic inquiry only in the mid-20th century. It is nowclear that there are different kinds of memory, which are supported by different brain systems.One major distinction can be drawn between working memory and long-term memory.Long-term memory can be separated into declarative (explicit) memory and a collection ofnondeclarative (implicit) forms of memory that include habits, skills, priming, and simpleforms of conditioning. These memory systems depend variously on the hippocampus andrelated structures in the parahippocampal gyrus, as well as on the amygdala, the striatum,cerebellum, and the neocortex. This work recounts the discovery of declarative and non-declarative memory and then describes the nature of declarative memory, working memory,nondeclarative memory, and the relationship between memory systems.

The idea that memory is not a single facultyhas a long history. In his Principles of Psychol-

ogy, William James (1890) wrote separate chap-ters on memory and habit. Bergson (1910) sim-ilarly distinguished between a kind of memorythat represents our past and memory that is notrepresentational but nevertheless allows the ef-fect of the past to persist into the present. Onefinds other antecedents as well. McDougall(1923) wrote about explicit and implicit recog-nition memory, and Tolman (1948) proposedthat there is more than one kind of learning.These early proposals were often expressed as adichotomy involving two forms of memory. Theterminologies differed, but the ideas were simi-lar. Thus, Ryle (1949) distinguished betweenknowing how and knowing that, and Bruner(1969) identified memory without record and

memory with record. Later, the artificial intelli-gence literature introduced a distinction be-tween declarative and procedural knowledge(Winograd 1975). Yet constructs founded inphilosophy and psychology are often abstractand have an uncertain connection to biology,that is, to how the brain actually stores informa-tion. History shows that as biological informa-tion becomes available about structure andfunction, understanding becomes more con-crete and less dependent on terminology.

THE DISCOVERY OF DECLARATIVE ANDNONDECLARATIVE MEMORY SYSTEMS

Biological and experimental inquiry into thesematters began with studies of the noted patientH.M. (Scoville and Milner 1957; Squire 2009).

Editors: Eric R. Kandel, Yadin Dudai, and Mark R. Mayford

Additional Perspectives on Learning and Memory available at www.cshperspectives.org

Copyright # 2015 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a021667

Cite this article as Cold Spring Harb Perspect Biol 2015;7:a021667

1

Harbor Laboratory Press at SERIALS/BIOMED LIB0175B on March 3, 2015 - Published by Cold Springhttp://cshperspectives.cshlp.org/Downloaded from

mailto:[email protected]

mailto:[email protected]

http://www.cshperspectives.org



http://cshperspectives.cshlp.org/

H.M. developed profound memory impair-ment following a bilateral resection of the me-dial temporal lobe, which had been performedto relieve severe epilepsy. The resection includedmuch of the hippocampus and the adjacentparahippocampal gyrus (Annese et al. 2014;Augustinack et al. 2014). H.M.’s memory im-pairment was disabling and affected all mannerof material (scenes, words, faces, etc.), so it wasquite unexpected when he proved capable oflearning a hand–eye coordination skill (mirrordrawing) over a period of 3 days (Milner 1962).He learned rapidly and efficiently but on eachtest day had no memory of having practiced thetask before. This finding showed that memory isnot a single entity. Yet, at the time, discussiontended to set aside motor skills as an exception,a less cognitive form of memory. The view wasthat all of the rest of memory was impaired inH.M. and that the rest of memory is of onepiece.

There were early suggestions in the animalliterature that more than just motor skills wereintact after lesions of hippocampus or relatedstructures (Gaffan 1974; Hirsh 1974; O’Keefeand Nadel 1978). However, these proposals dif-fered from each other, and they came at a timewhen the findings in experimental animals didnot conform well to the findings for humanmemory and amnesia. In particular, animalswith hippocampal lesions often succeeded attasks that were failed by patients with similarlesions. It gradually became clear that animalsand humans can approach the same task withdifferent strategies (and using different brainsystems), and also that patients with medial tem-poral lobe lesions, like experimental animalswith similar lesions, can in fact succeed at awide range of learning and memory abilities.

First came the finding that memory-im-paired patients could acquire, at a normal rate,the perceptual skill of reading mirror-reversedwords, despite poor memory for the task andfor the words that were read (Cohen and Squire1980). Thus, perceptual skills, not just motorskills, were intact. This finding was presentedin the framework of a brain-based distinctionbetween two major forms of memory that af-ford either declarative or procedural knowledge.

Declarative knowledge is knowledge available asconscious recollection, and it can be brought tomind as remembered verbal or nonverbal ma-terial, such as an idea, sound, image, sensation,odor, or word. Procedural knowledge refers toskill-based information. What is learned is em-bedded in acquired procedures and is expressedthrough performance.

Subsequently, other forms of experience-dependent behaviors were found to be distinctfrom declarative memory. One important phe-nomenon was priming—the improved abilityto detect, produce, or classify an item basedon a recent encounter with the same or relateditem (Tulving and Schacter 1990; Schacter andBuckner 1998). For example, individuals willname objects faster on their second presenta-tion, and independent of whether they recog-nize the objects as having been presented before.

Another important discovery was that theneostriatum (not the medial temporal lobe) isimportant for the sort of gradual feedback-guided learning that results in habit memory(Mishkin et al. 1984; Packard et al. 1989; Knowl-ton et al. 1996). Tasks that assess habit learningare often structured so that explicit memoriza-tion is not useful (e.g., because the outcome ofeach trial is determined probabilistically), andindividuals must depend more on a gut feeling.After learning, it is more accurate to say thatindividuals have acquired a disposition to per-form in a particular way than to say that theyhave acquired a fact (i.e., declarative knowledge)about the world.

Given the wide variety of learning andmemory phenomena that could eventually beshown in patients (e.g., priming and habitlearning as well as simple forms of classical con-ditioning), the perspective eventually shifted toa framework involving multiple memory sys-tems rather than just two kinds of memory(Fig. 1). Accordingly, the term “nondeclarative”was introduced with the idea that nondeclara-tive memory is an umbrella term referring tomultiple forms of memory that are not declar-ative (Squire and Zola-Morgan 1988). Nonde-clarative memory includes skills and habits,simple forms of conditioning, priming, andperceptual learning, as well as phylogenetically

L.R. Squire and A.J.O. Dede

2 Cite this article as Cold Spring Harb Perspect Biol 2015;7:a021667



early forms of behavioral plasticity like habitu-ation and sensitization that are well developedin invertebrates. The various memory systemscan be distinguished in terms of the differentkinds of information they process and the prin-ciples by which they operate.

Declarative memory provides a way to rep-resent the external world. It is the kind of mem-ory we typically have in mind when we use theterm memory in everyday language. Declarativememory has two major components, semanticmemory (facts about the world) and episodicmemory (the ability to re-experience a time-and-place-specific event in its original context)(Tulving 1983). The acquisition of episodicmemory requires the involvement of brain sys-tems in addition to medial temporal lobe struc-tures, especially the frontal lobes (Tulving 1989;Shimamura et al. 1991). There is some uncer-tainty around the issue of whether nonhumananimals have the capacity for episodic memory(i.e., the capacity for mental time travel that canreturn an animal to the scene of an earlierevent), and the idea has been difficult to putto the test (Tulving 2005). For elegant demon-strations of episodic-like memory in nonhumananimals, see Clayton and Dickinson (1998).

Nondeclarative memory is dispositional andis expressed through performance rather thanrecollection. An important principle is the abil-

ity to gradually extract the common elementsfrom a series of separate events. Nondeclarativememory provides for myriad unconscious waysof responding to the world. The unconsciousstatus of nondeclarative memory creates someof the mystery of human experience. Here arisethe habits and preferences that are inaccessibleto conscious recollection, but they neverthelessare shaped by past events, they influence ourcurrent behavior and mental life, and they area fundamental part of who we are.

Sherry and Schacter (1987) suggested thatmultiple memory systems evolved because theyserve distinct and fundamentally different pur-poses. For example, the gradual changes thatoccur in birdsong learning are different from,and have a different function than, the rapidlearning that occurs when a bird caches foodfor later recovery. These memory systems oper-ate in parallel to support and guide behavior. Forexample, imagine an unpleasant event from ear-ly childhood, such as being knocked down by alarge dog. Two independent consequences of theevent could potentially persist into adulthood asdeclarative and nondeclarative memories. Onthe one hand, the individual might have a con-scious, declarative memory of the event itself.On the other hand, the individual might havea fear of large dogs, quite independently ofwhether the event itself is remembered. Note

Memory

Nondeclarative

Simpleclassical

conditioning

Nonassociativelearning

Reflexpathways

Cerebellum

Skeletalresponses

Emotionalresponses

AmygdalaNeocortexStriatumMedial temporal lobediencephalon

Declarative

Facts Events Procedural(skills and

habits)

Primingand

perceptuallearning

Figure 1. Organization of mammalian long-term memory systems. The figure lists the brain structures thoughtto be especially important for each form of memory. In addition to its central role in emotional learning, theamygdala is able to modulate the strength of both declarative and nondeclarative memory.


Cite this article as Cold Spring Harb Perspect Biol 2015;7:a021667 3



that a fear of dogs would not be experienced as amemory but rather as a part of personality, apreference, or an attitude about the world.

THE NATURE OF DECLARATIVE MEMORY

Studies of patients and experimental animalswith medial temporal lobe damage have identi-fied four task requirements that reliably revealimpaired memory: (1) tasks where learning oc-curs in a single trial or single study episode(Mishkin 1978); (2) tasks where associationsbetween stimuli are learned across space andtime (e.g., Higuchi and Miyashita 1996; Fortinet al. 2002); (3) tasks where the acquired infor-mation can be used flexibly (e.g., Bunsey andEichenbaum 1996; Smith and Squire 2005); and(4) tasks where learning depends on awarenessof what is being learned (Clark and Squire 1998;Smith and Squire 2008).

Declarative memory (sometimes termedexplicit memory) is well adapted for the rapidlearning of specific events. Declarative memory

allows remembered material to be comparedand contrasted. The stored representations areflexible, accessible to awareness, and can guideperformance in multiple different contexts.The key structures that support declarative mem-ory are the hippocampus and the adjacententorhinal, perirhinal, and parahippocampalcortices, which make up much of the parahip-pocampal gyrus (Fig. 2) (Squire and Zola-Mor-gan 1991). These structures are organized hier-archically, and their anatomy suggests how thestructures might contribute differently to theformation of declarative memory, for example,in the encoding of objects (perirhinal cortex)or scenes (parahippocampal cortex) and in theforming of associations between them (hippo-campus) (Squire et al. 2004; Davachi 2006;Staresina et al. 2011).

Structures in the diencephalic midline(mammillary nuclei, medial dorsal nucleus, an-terior thalamic nuclei, together with the internalmedullary lamina and the mammillothalamictract) are also important for declarative memo-

Other directprojections

Hippocampus

Dentategyrus

CA3

CA1

Subicularcomplex

Entorhinalcortex (EC)

Perirhinal cortex(PR)

Parahippocampalcortex (PC)

Unimodal and polymodal association areas(frontal, temporal, and parietal lobes)

Figure 2. Schematic view of the medial temporal lobe memory system for declarative memory, which is com-posed of the hippocampus and the perirhinal, entorhinal, and parahippocampal cortices. In addition to theconnections shown here, there are also weak projections from the perirhinal and parahippocampal cortices tothe CA1-subiculum border.





ry. Damage to these structures causes the samecore deficit as damage to the medial temporallobe, probably because these nuclei and tractsare anatomically related to the medial temporallobe (see Markowitsch 1988; Victor et al. 1989;Harding et al. 2000; Squire and Wixted 2011).

Discussion continues about the nature ofdeclarative memory and about when exactlythe hippocampus (and related structures) is in-volved in learning and memory. One proposal isthat, whereas conscious recollection depends onthe hippocampus as described above, the hip-pocampus is also important for unconsciousmemory under some circumstances (Henke2010; Hannula and Greene 2012; Shohamyand Turk-Browne 2013). One way to explorethis issue has been to record eye movementswhile volunteers are making behavioral memo-ry judgments. In some situations (e.g., when achange has occurred in the layout of a scene),the eyes do not reveal signs of memory (by mov-ing to the changed location) unless participantsare aware of where the change occurred (Smithet al. 2008). Furthermore, these eye movementeffects require the integrity of the hippocampus.Nevertheless, in other situations, eye move-ments can signal which item is correct, and cor-relate with hippocampal activity, even when be-havioral memory judgments are incorrect andparticipants are therefore thought to be un-aware (Hannula and Ranganath 2009). Such afinding could mean that eye movements (andhippocampal activity) can index unaware mem-ory. Yet, it is also true that awareness is (pre-sumably) continuous, and a low amount ofawareness is not the same as a complete lack ofawareness (Kumaran and Wagner 2009). Just asrecognition memory can succeed when free re-call fails, eye movements might reveal signs ofaware memory when recognition fails. It wouldbe instructive in this circumstance to obtainconfidence ratings in association with memoryjudgments and ask whether there is any detect-able awareness of which items are correct.

Other work has implicated medial tempo-ral lobe structures in the unaware learning ofsequences and other tasks with complex con-tingencies (Chun and Phelps 1999; Rose et al.2002; Schendan et al. 2003). This idea is often

based on functional magnetic resonance imag-ing (fMRI) evidence of medial temporal lobeactivity during unaware learning. Strategic fac-tors (e.g., explicit attempts to memorize) mayexplain some of these effects (Westerberg et al.2011), together with the possibility that aware-ness may not be entirely absent (Poldrack andRodriguez 2003). In addition, considering thatfMRI data cannot establish a necessary role for aparticular structure, it will be useful to supple-ment fMRI datawith evidence that patients withhippocampal lesions are impaired at the sametasks that afford unaware learning. Interestingly,for some of these tasks, hippocampal patientswere not impaired (Reber and Squire 1994;Manns and Squire 2001), but impairment hasbeen reported in patients when the damage wasundescribed or extended beyond the hippocam-pus (Chun and Phelps 1999). Studies that com-bine fMRI, patient data, and rigorous measuresof awareness will be useful in pursuing this in-teresting issue.

WORKING MEMORY

Working memory refers to the capacity to main-tain a limited amount of information in mind,which can then support various cognitive abili-ties, including learning and reasoning (Baddeley2003). Within cognitive neuroscience, the term“working memory” has largely replaced the lessprecise term “short-term memory.” (Note thatthe term “short-term memory” remains usefulin cellular neuroscience where it has a differentand distinct meaning [Kandel et al. 2014].) His-torically, working memory has been consideredto be distinct from long-term memory and in-dependent of the medial temporal lobe struc-tures that support the formation of long-termmemory (Drachman and Arbit 1966; Atkinsonand Shiffrin 1968; Baddeley and Warrington1970; Milner 1972). Long-term memory isneeded when the capacity of working memoryis exceeded or whenworking memory is disrupt-ed by diverting attention to different material.

Uncertainty can arise when determining inany particular case, whether performance de-pends on working memory or whether somuch information needs to be kept in mind





that working memory capacity is exceeded andperformance must rely on long-term memory.What is the situation when patients fail at taskswith short retention intervals, or no retentioninterval (Hannula et al. 2006; Olson et al. 2006;Warren et al. 2011, 2012; Yee et al. 2014)? Arethese long-term memory tasks or, as has beensuggested, do such findings show that workingmemory sometimes depends on the medialtemporal lobe (Ranganath and Blumenfeld2005; Graham et al. 2010). It is important tonote that working memory cannot be definedin terms of any particular retention interval (Je-neson and Squire 2012). Even when the reten-tion interval is measured in seconds, workingmemory capacity can be exceeded such that per-formance must depend, in part, on long-termmemory (e.g., immediately after presentation ofa list of 10 words).

Findings from a scene-location task illus-trate the problem. The task involved scenes con-taining a number of different objects. On eachtrial, a scene was presented together with a ques-tion (e.g., is the plant on the table?). A few sec-onds later, the same scene was presented againbut now the object queried about might ormight not appear in a different location. In thiscondition, patients with hippocampal lesionswere accurate at detecting whether or not theobject had moved (Jeneson et al. 2011). How-ever, when attention was drawn to four differentobjects, any one of which might move, patientswere impaired (Yee et al. 2014). It is likely thatthe impairment in the second condition oc-curred because working memory capacity wasexceeded. Visual working memory capacity isquite limited and, typically, even healthy youngadults can maintain only three to four simplevisual objects in working memory (Cowan2001; Fukuda et al. 2010).

In any case, there are ways to distinguishworking memory and long-term memory(Shrager et al. 2008; Jeneson et al. 2010). Forexample, one can vary the number of items orassociations to be remembered and ask whetherpatients show a sharp discontinuity in perfor-mance as the number of items increases andworking memory capacity is exceeded. In onestudy (Jeneson et al. 2010), patients with hip-

pocampal lesions or large medial temporal lobelesions saw different numbers of objects (1 to 7)on a tabletop and then immediately tried toreproduce the array on an adjacent table. Per-formance was intact when only a few object lo-cations needed to be remembered. However,there was an abrupt discontinuity in perfor-mance with larger numbers of object locations.One patient (who had large medial temporallobe lesions similar to H.M.) learned one-,two-, or three-object locations as quickly asdid controls, never needing more than one ortwo trials to succeed. Yet when four-object lo-cations needed to be remembered, he did notsucceed even after 10 trials with the same array.These findings indicate that the ability to main-tain small numbers of object–place associationsin memory is intact after medial temporal lobelesions. An impairment was evident only whena capacity limit was reached, at which pointperformance needed to depend on long-termmemory. A similar conclusion was reached instudies of a single patient with restricted hippo-campal lesions. Performance was fully intacton an extensive working memory battery, in-cluding tasks of relational (associative) memory(Baddeley et al. 2010, 2011).

Spatial tasks like path integration can also beperformed normally by patients with medialtemporal lobe lesions, as long as the task canbe managed within working memory (Shrageret al. 2008; Kim et al. 2013). The findings aredifferent for rats with hippocampal lesions(Kim et al. 2013), either because for the ratthe task relies on long-term memory or becausethe rat hippocampus is needed for some onlinespatial computations, as suggested previously(Whitlock et al. 2008).

fMRI findings are relevant to these issues,because medial temporal lobe activity is some-times found in association with short-delay rec-ognition memory tasks (Ranganath and D’Es-posito 2001; Piekema et al. 2006; Toepper et al.2010). Yet, it is noteworthy that the extent ofmedial temporal lobe activity in short-delaytasks can be modulated by memory demands.For example, in some studies, the medial tem-poral lobe activity that occurred while main-taining information in memory was correlated





with subsequent retention of the material beinglearned (Schon et al. 2004; Ranganath et al.2005; Nichols et al. 2006). In addition, in astudy that required maintaining faces in mem-ory, the connectivity between the hippocampusand the fusiform face area increased with highermnemonic load (one face vs. four faces) (Riss-man et al. 2008). Concurrently, with higherload, the connectivity decreased between frontalregions traditionally linked to working memory(Goldman-Rakic 1995; Postle 2006) and the fu-siform face area. These findings suggest thatfMRI activity in the medial temporal lobe re-flects processes related to the formation of long-term memory rather than processes related toworking memory itself (for review, see Jenesonand Squire 2012).

NONDECLARATIVE MEMORY

Nondeclarative memory (sometimes termedimplicit memory) refers to a collection of abil-ities that are expressed through performancewithout requiring conscious memory content.Study of nondeclarative memory began withmotor skills and perceptual skills, as describedabove, but soon included additional abilities aswell. The next of these to come under study wasthe phenomenon of priming. Priming is evi-dent as improved access to items that havebeen recently presented or improved access toassociates of those items. This improvement isunconscious and is experienced as part of per-ception, as perceptual fluency, not as an expres-sion of memory. A key finding was that primingeffects were intact in memory-impaired pa-tients. For example, patients can perform nor-mally on tests that use word stems as cues forrecently presented words (e.g., study BRICK,CRATE; test with BRI___, CRA___). Impor-tantly, performance was intact in patients onlywhen they were instructed to complete each cueto form the first word that comes to mind. Withconventional memory instructions (use thecue to help recall a recently presented word),healthy volunteers outperformed the patients(Graf et al. 1984). Priming can occur for mate-rial that has no preexisting memory representa-tion (e.g., nonsense letter strings) (Hamann and

Squire 1997a) and for material that is related bymeaning to recently studied items. Thus, whenasked to free associate to a word (e.g., strap),volunteers produced a related word (e.g., belt)more than twice as often when that word (belt)was presented recently (Levy et al. 2004). Im-portantly, severely amnesic patients showed ful-ly intact word priming, even while performingat chance levels in parallel memory tests (Ha-mann and Squire 1997b; Levy et al. 2004). Thus,priming occurs but it does not benefit con-scious memory decisions. Indeed, direct mea-surements showed that priming provides only aweak and unreliable cue for conscious judg-ments of familiarity (Conroy et al. 2005).

Priming is presumably advantageous be-cause animals evolved in a world where thingsthat are encountered once are likely to be en-countered again. Priming improves the speedand efficiency with which organisms interactwith a familiar environment and may influencefeature-based attentional processes (Hutchin-son and Turk-Browne 2012; Theeuwes 2013).Evoked-potential studies indicate that the elec-trophysiological signature of priming occursearly and well before the activity that signalsconscious recognition of a past event (Palleret al. 2003). In neuroimaging studies, primingis often associated with reduced activity in re-gions of neocortex relevant to the task (Squireet al. 1992; Schacter et al. 2007). A similar find-ing (repetition suppression) (Desimone 1996)has been described in nonhuman primates (astimulus-specific attenuation in firing ratewith repeated presentation of a stimulus), andmay underlie the phenomenon of priming(Wiggs and Martin 1998). Models have beenproposed to explain how a net reduction in cor-tical activity could allow for faster perceptualprocessing (i.e., priming) (Grill-Spector et al.2006). Some studies have found a correlationbetween behavioral measures of priming andreduced activity in the prefrontal cortex (Mac-cotta and Buckner 2004). This result has notbeen found in ventral temporal cortex for eitherhumans or nonhuman primates (Maccotta andBuckner 2004; McMahon and Olson 2007).

Changes in cortex also underlie the relatedphenomenon of perceptual learning (Gilbert





et al. 2009). Perceptual learning refers to gradualimprovement in the detection or discriminationof visual stimuli with repeated practice. Chang-es in cortical circuitry during perceptual learn-ing are detectable as early as primary visual cor-tex (V1) and may depend in part on structuralchanges in the long-range horizontal connec-tions formed by V1 pyramidal cells (Gilbertand Li 2012). This circuitry is under the controlof bottom-up processes as well as top-down in-fluences related to attention and behavioralcontext (Gilbert and Li 2013).

Another early example of nondeclarativememory was simple classical conditioning,best illustrated in the literature of delay eyeblinkconditioning. In delay conditioning, a neutralconditioned stimulus (CS), such as a tone, ispresented just before an unconditioned stimu-lus (US), such as an airpuff to the eye. The twostimuli then overlap and coterminate. Critically,delay eyeblink conditioning is intact in amnesiaand is acquired independently of awareness(Gabrieli et al. 1995; Clark and Squire 1998).Participants who did not become aware of therelationship between the CS and US (i.e., thatthe CS predicts the US) learned just as well asvolunteers who did become aware (Manns etal. 2001). Indeed, when CS–US associationstrength was varied (by changing the numberof consecutive CS alone or CS–US presenta-tions), the probability of a conditioned responseincreased with association strength but was in-versely related to how much the US was expect-ed (Clark et al. 2001). Largely on the basis ofwork with rabbits, delay eyeblink conditioningproved to depend on the cerebellum and asso-ciated brain stem circuitry (Thompson andKrupa 1994; Thompson and Steinmetz 2009).Forebrain structures are not necessary for acqui-sition or retention of classically conditionedeyeblink responses.

Evaluative information, that is, whether astimulus has positive or negative value, is ac-quired largely as nondeclarative memory. Bio-logical study of this kind of memory has focusedespecially on the associative learning of fear(Davis 2006; Adolphs 2013; LeDoux 2014). Itsnondeclarative status is illustrated by the factthat, in humans, associative fear learning pro-

ceeded normally after hippocampal lesions,even though the CS–US pairings could not bereported (Bechara et al. 1995). The amygdalahas a critical role in fear learning, and its func-tion (as well as its connectivity) appears to beconserved widely across species. In human neu-roimaging studies, the amygdala was activatednot only by fear but by strongly positive emo-tions as well (Hamann et al. 2002). Thus, theamygdala appears to be critical for associatingsensory stimuli with stimulus valence. Ordinar-ily, animals express fear learning by freezing be-havior (immobility). However, in a task wherelearned fear must instead be expressed by exe-cuting an avoidance response (an escape), freez-ing is maladaptive. In this case, prefrontal cor-tex inhibits defense behaviors (such as freezing)that are mediated by the amygdala, thereby al-lowing the animal to escape (Moscarello andLeDoux 2013). Inhibitory action of the prefron-tal cortex on the amygdala (from infralimbicprefrontal cortex in rat or from ventromedialprefrontal cortex in humans) has also beenfound to occur during the reversal of fear learn-ing (i.e., extinction) (Milad and Quirk 2012).This work has relevance for clinical disorders,such as phobias and posttraumatic stress disor-der (Davis 2011).

In addition to these functions, it is impor-tant to note that the amygdala also exerts a mod-ulatory influence on both declarative and non-declarative memory. This role of the amygdala isthe basis for the fact that emotionally arousingevents are typically remembered better thanemotionally neutral events. The mechanismfor this effect is understood and depends onthe release of stress hormones from the adrenalgland, which affects the forebrain via the vagusnerve, the nucleus of the solitary tract, and thelocus coeruleus. Ultimately, the effect is medi-ated by the amygdala through its basolateral nu-cleus (McGaugh and Roozendaal 2009).

The gradual trial-and-error learning thatleads to the formation of habits was proposedin the 1980s to be supported by the striatum(Mishkin et al. 1984), and habit memory sub-sequently became an important focus of study(Yin and Knowlton 2006; Graybiel 2008; Lilje-holm and O’Doherty 2012). Habit memory is





characterized by automatized, repetitive behav-ior and, unlike declarative memory, is insensi-tive to changes in reward value (Dickinson1985). An early demonstration of the distinc-tion between declarative memory and habitmemory came from rats with fornix lesions orcaudate lesions tested on two ostensibly similartasks. Rats with fornix lesions, which disrupthippocampal function, failed when they neededto acquire a flexible behavior but succeededwhen they needed to respond repetitively. Ratswith caudate lesions showed the opposite pat-tern (Packard et al. 1989). A similar contrastbetween declarative memory and habit memorywas shown for memory-impaired patients withhippocampal lesions and patients with nigro-striatal damage caused by Parkinson’s disease(Knowlton et al. 1996). In the task, probabilisticclassification, participants gradually learnedwhich of two outcomes (sun or rain) wouldoccur on each trial, given the particular combi-nation of four cues that appeared. One, two, orthree cues could appear on any trial, and eachcue was independently and probabilistically re-lated to the outcome. Patients with hippocam-pal lesions learned the task at a normal rate butcould not report facts about the task. Parkinsonpatients remembered the facts but could notlearn the task.

Tasks that can be learned quickly by memo-rization can also be learned by a trial-and-error,habit-based strategy, albeit much more slowly.In one study, healthy volunteers were able tolearn eight separate pairs of “junk objects” with-in a single session of 40 trials (i.e., choose thecorrect object in each pair). Two severely amne-sic patients with large medial temporal lobe le-sions also learned but only gradually, requiringmore than 25 test sessions and 1000 trials (Bay-ley et al. 2005). Unlike declarative memory,which is flexible and can guide behavior in dif-ferent contexts, the acquired knowledge in thiscase was rigidly organized. Patient performancecollapsed when the task format was altered byasking participants to sort the 16 objects intocorrect and incorrect groups (a trivial task forcontrols). In addition, although by the end oftraining the patients were consistently perform-ing at a high level, at the start of each test day

they were never able to describe the task, theinstructions, or the objects. Indeed, duringtesting they expressed surprise that they wereperforming so well. These findings provide par-ticularly strong evidence for the distinctionbetween declarative (conscious) and nondeclar-ative (unconscious) memory systems.

Reward-based learning of this kind dependson dopamine neurons in the midbrain (substan-tia nigra and ventral tegmental area), which pro-ject to the striatum and signal the informationvalue of the reward (Schultz 2013). The dorso-lateral striatum is crucial for the development ofhabits in coordination with other brain regions.Neurophysiological studies in mice during skilllearning documented that the dorsolateral stri-atum was increasingly engaged as performancebecame more automatic and habit-like (Yinet al. 2009). In contrast, the dorsomedial stria-tum was engaged only early in training. Similar-ly, in rats learning a conditioned T-maze task,activity gradually increased in the dorsolateralstriatum as training progressed, and this activi-ty correlated with performance (Thorn et al.2010). In the dorsomedial striatum, activity firstincreased but then decreased as training pro-gressed. Increased activity in the dorsolateralstriatum during the later stages of habit forma-tion occurred together with late-developing ac-tivity in infralimbic cortex (Smith and Graybiel2013). Moreover, disruption of infralimbic cor-tex during late training prevented habit forma-tion. Thus, these two regions (dorsolateral stri-atum and infralimbic cortex) appear to worktogether to support a fully formed habit.

THE RELATIONSHIP BETWEEN MEMORYSYSTEMS

The memory systems of the mammalian brainoperate independently and in parallel to sup-port behavior, and how one system or anothergains control is a topic of considerable interest(Poldrack and Packard 2003; McDonald andHong 2013; Packard and Goodman 2013). Insome circumstances, memory systems are de-scribed as working cooperatively to optimizebehavior and in other circumstances are de-scribed as working competitively. However, it





is not easy to pin down what should count for oragainst cooperativity, competition, or indepen-dence in any particular case. For example, thefact that the manipulation of one memory sys-tem can affect the operation of another has beentaken as evidence for competition between sys-tems (Schwabe 2013). Yet, even for systems thatare strictly independent, the loss of one systemwould be expected to affect the operation ofanother system by affording it more opportuni-ty to control behavior.

Much of the experimental work on the rela-tionship between memory systems has focusedon hippocampus-dependent declarative mem-ory and dorsolateral striatum-dependent habitmemory. In an illustrative study (Packard andMcGaugh 1996), rats were trained in a four-arm, plus-shaped maze to go left, always begin-ning in the south arm (and with the north armblocked). In this situation, rats could learn ei-ther a place (the left arm) or a response (turnleft). To discriminate between these two possi-bilities, rats were occasionally started in thenorth arm (with the south arm now blocked).When these north-arm starts were given early intraining, rats tended to enter the same arm thathad been rewarded. They had learned a place.However, with extended training, rats tended torepeat the learned left-turn response and enterthe unrewarded arm. Place learning was abol-ished early in training by lidocaine infusionsinto the hippocampus. In this case, rats showedno preference for either arm. Correspondingly,later in training, response learning was abol-ished by lidocaine infusions into the caudatenucleus. In this case, however, rats showed placeresponding. In other words, even though be-havior later in training was guided by caudate-dependent response learning, information re-mained available about place. When the caudatenucleus was inactive, the parallel memory sys-tem supported by the hippocampus was un-masked.

A similar circumstance has been describedin humans performing a virtual navigation taskthat could be solved by either a spatial or non-spatial (habit-like) strategy (Iaria et al. 2003). Atthe outset, spatial and nonspatial strategies wereadopted equally often, but as training prog-

ressed participants tended to shift to a nonspa-tial strategy. Participants who used the spatialstrategy (navigating in relation to landmarks)showed increased activity in the right hippo-campus early in training. Participants usingthe nonspatial strategy (counting maze arms)showed increased activity in the caudate nucle-us, which emerged as training progressed.

Several factors increase the tendency toadopt a striatal strategy, including stress (Kimet al. 2001; Schwabe 2013), psychopathology(Wilkins et al. 2013), aging (Konishi et al.2013), and a history of alcohol and drug use(Bohbot et al. 2013). Prefrontal cortex mayalso be important in determining which mem-ory system gains control over behavior (Mc-Donald and Hong 2013).

Although many tasks can be acquired bymore than one memory system, other tasksstrongly favor one system over another. In thiscircumstance, engaging the less optimal systemcan interfere with performance. Thus, fornixlesions in rats facilitated acquisition of a cau-date-dependent maze habit that required re-peated visits to designated arms (Packard et al.1989). The fornix lesion presumably disruptedthe tendency to use a nonoptimal declarativememory strategy. Similarly, a familiar featureof skill learning in humans is that trying tomemorize, and use declarative memory, candisrupt performance.

Neuroimaging studies show that feedback-guided learning typically engages the striatum.A task described earlier, probabilistic classifica-tion, requires participants to make a guess oneach trial based on cues that are only partiallyreliable. Participants typically begin by trying tomemorize the task structure but then turn to thehabit-like strategy of accumulating responsestrength in association with the cues. Corre-spondingly, fMRI revealed activity in the medialtemporal lobe early during learning (Poldrackand Gabrieli 2001). As learning progressed, ac-tivity decreased in the medial temporal lobe,and activity increased in the striatum. Moreover,when the task was modified so as to encouragethe use of declarative memory, less activity wasobserved in the striatum and more activity wasobserved in the medial temporal lobe.





CONCLUSION

The memory system framework is fundamentalto the contemporary study of learning andmemory. Within this framework, the variousmemory systems have distinct purposes and dis-tinct anatomy, and different species can solvethe same task using different systems. Interest-ingly, efforts have been made to account forsome findings (e.g., priming or classificationlearning) with models based on a single system(Zaki et al. 2003; Zaki 2004; Berry et al. 2012).Yet, these accounts have difficulty explainingdouble dissociations (e.g., Packard et al. 1989;Knowlton et al. 1996), chance performance ontests of declarative memory when priming isintact (Hamann and Squire 1997b), and suc-cessful habit learning in the face of expressedignorance about the task (Bayley et al. 2005).

One implication of these facts is that thetherapeutic targets for various kinds of memorydisorders are quite different. For example, forextreme fear-based memories like phobias, onemust target the amygdala, for strong habit-based memories like obsessive–compulsive dis-orders, one must target the striatum, and forsevere forgetfulness, as in Alzheimer’s disease,one must target the hippocampus and adjacentstructures. The notion of multiple memory sys-tems is now widely accepted and establishes animportant organizing principle across speciesfor investigations of the biology of memory.

REFERENCES

Adolphs R. 2013. The biology of fear. Curr Biol 23: R79–R93.

Annese J, Schenker-Ahmed NM, Bartsch H, Maechler P,Sheh C, Thomas N, Kayano J, Ghatan A, Bresler N, FroschMP, et al. 2014. Postmortem examination of patientH.M.’s brain based on histological sectioning and digital3D reconstruction. Nat Commun 5: 3122.

Atkinson RC, Shiffrin RM. 1968. Human memory: A pro-posed system and its control processes. In Psychology oflearning and motivation: Advances in research and theory(ed. Spence KW, Spence JT), pp. 89–195. Academic,New York.

Augustinack JC, van der Kouwe AJW, Salat DH, Benner T,Stevens AA, Annese J, Fischl B, Frosch MP, Corkin S.2014. H.M.’s contributions to neuroscience: A reviewand autopsy studies. Hippocampus 24: 1267–1286.

Baddeley A. 2003. Double dissociations: Not magic, but stilluseful. Cortex 39: 129–131.

Baddeley AD, Warrington EK. 1970. Amnesia and the dis-tinction between long- and short-term memory. J VerbLearn Verb Behav 9: 176–189.

Baddeley A, Allen R, Vargha-Khadem F. 2010. Is the hippo-campus necessary for visual and verbal binding in work-ing memory? Neuropsychologia 48: 1089–1095.

Baddeley A, Jarrold C, Vargha-Khadem F. 2011. Workingmemory and the hippocampus. J Cog Neurosci 23:3855–3861.

Bayley PJ, Frascino JC, Squire LR. 2005. Robust habit learn-ing in the absence of awareness and independent of themedial temporal lobe. Nature 436: 550–553.

Bechara A, Tranel D, Damasio H, Adolphs R, Rockland C,Damasio AR. 1995. Double dissociation of conditioningand declarative knowledge relative to the amygdala andhippocampus in humans. Science 269: 1115–1118.

Bergson HL. 1910. Matter and memory (trans. Paul NM,Palmer WS). George Allen, London.

Berry CJ, Shanks DR, Speekenbrink M, Henson RNA. 2012.Models of recognition, repetition priming, and fluency:Exploring a new framework. Psych Rev 119: 40–79.

Bohbot VD, Del Balso D, Conrad K, Konishi K, Leyton M.2013. Caudate nucleus-dependent navigational strategiesare associated with increased use of addictive drugs. Hip-pocampus 23: 973–984.

Bruner JS. 1969. Modalities of memory. In The pathology ofmemory (ed. Talland GA, et al.). pp. 253–259. Academic,New York.

Bunsey M, Eichenbaum H. 1996. Conservation of hippo-campal memory function in rats and humans. Nature379: 255–257.

Chun MM, Phelps EA. 1999. Memory deficits for implicitcontextual information in amnesic subjects with hippo-campal damage. Nat Neurosci 2: 844–847.

Clark RE, Squire LR. 1998. Classical conditioning and brainsystems: A key role for awareness. Science 280: 77–81.

Clark RE, Manns JR, Squire LR. 2001. Trace and delay eye-blink conditioning: Contrasting phenomena of declara-tive and nondeclarative memory. Psychol Sci 12: 304–308.

Clayton RS, Dickinson A. 1998. Episodic-like memory dur-ing cache recovery by scrub jays. Nature 395: 272–274.

Cohen NJ, Squire LR. 1980. Preserved learning and reten-tion of pattern analyzing skill in amnesia: Dissociation ofknowing how and knowing that. Science 210: 207–209.

Conroy MA, Hopkins RO, Squire LR. 2005. On the contri-bution of perceptual fluency and priming to recognitionmemory. Cogn Affect Behav Neurosci 5: 14–20.

Cowan N. 2001. The magical number 4 in short-term mem-ory: A reconsideration of mental storage capacity. BehavBrain Sci 24: 87–185.

Davachi L. 2006. Item, context and relational episodic en-coding in humans. Curr Opin Neurobiol 16: 693–700.

Davis M. 2006. Neural systems involved in fear and anxietymeasured with fear-potentiated startle. Am Psychol 61:741–756.

Davis M. 2011. NMDA receptors and fear extinction: Im-plications for cognitive behavioral therapy. DialoguesClin Neurosci 13: 463–474.





Desimone R. 1996. Neural mechanisms for visual memoryand their role in attention. Proc Natl Acad Sci 93: 13494–13499.

Dickinson A. 1985. Actions and habits: the development ofbehavioural autonomy. Phil Trans R Soc Lond B Biol Sci308: 67–78.

Drachman DA, Arbit J. 1966. Memory and the hippocampalcomplex. II. Is memory a multiple process? Arch Neurol15: 52–61.

Fortin NJ, Agster KL, Eichenbaum HB. 2002. Critical role ofthe hippocampus in memory for sequences of events. NatNeurosci 5: 458–462.

Fukuda K, Awh E, Vogel EK. 2010. Discrete capacity limits invisual working memory. Curr Opin Neurobiol 20: 177–182.

Gabrieli JDE, McGlinchey-Berroth R, Carrillo MC, GluckMA, Cermak LS, Disterhoft JF. 1995. Intact delay-eye-blink classical conditioning in amnesia. Behav Neurosci109: 819–827.

Gaffan D. 1974. Recognition impaired and association intactin the memory of monkeys after transaction of the fornix.J Comp Physiol Psychol 86: 1100–1109.

Gilbert CD, Li W. 2012. Adult visual cortical plasticity.Neuron 75: 250–264.

Gilbert CD, Li W. 2013. Top-down influences on visual pro-cessing. Nat Rev Neurosci 14: 350–363.

Gilbert CD, Li W, Piech V. 2009. Perceptual learning andadult cortical plasticity. J Physiol 587: 2743–2751.

Goldman-Rakic PS. 1995. Architecture of the prefrontalcortex and the central executive. Ann NY Acad Sci 769:71–83.

Graf P, Squire LR, Mandler G. 1984. The information thatamnesic patients do not forget. J Exp Psychol Learn MemCog 10: 164–178.

Graham KS, Barense MD, Lee AC. 2010. Going beyond LTMin the MTL: A synthesis of neuropsychological and neu-roimaging findings on the role of the medial temporallobe in memory and perception. Neuropsychologia 48:831–853.

Graybiel AM. 2008. Habits, rituals, and the evaluative brain.Ann Rev Neurosci 31: 359–387.

Grill-Spector K, Henson R, Martin A. 2006. Repetition andthe brain: Neural models of stimulus-specific effects.Trends Cognit Sci 10: 14–23.

Hamann SB, Squire LR. 1997a. Intact priming for novelperceptual representations in amnesia. J Cog Neurosci 9:699–713.

Hamann SB, Squire LR. 1997b. Intact perceptual memory inthe absence of conscious memory. Behav Neurosci 111:850–854.

Hamann SB, Ely TD, Hoffman JM, Kilts CD. 2002. Ecstasyand agony: Activation of the human amygdala in positiveand negative emotion. Psychol Sci 13: 135–141.

Hannula DE, Greene AJ. 2012. The hippocampus reevalu-ated in unconscious learning and memory: At a tippingpoint? Front Human Neurosci 6: 80.

Hannula DE, Ranganath C. 2009. The eyes have it: Hippo-campal activity predicts expression of memory in eyemovements. Neuron 63: 592–599.

Hannula DE, Tranel D, Cohen NJ. 2006. The long and theshort of it: Relational memory impairments in amnesia,even at short lags. J Neurosci 26: 8352–8359.

Harding A, Halliday G, Caine D, Krill J. 2000. Degeneratonof anterior thalamic nuclei differentiates alcoholics withamnesia. Brain 123: 141–154.

Henke K. 2010. A model for memory systems based onprocessing modes rather than consciousness. Nat RevNeurosci 11: 523–532.

Higuchi S, Miyashita Y. 1996. Formation of mnemonic neu-ronal responses to visual paired associates in inferotem-poral cortex is impaired by perirhinal and entorhinallesions. Proc Natl Acad Sci 93: 739–743.

Hirsh R. 1974. The hippocampus and contextual retrieval ofinformation from memory: A theory. Behav Biol 12:421–444.

Hutchinson JB, Turk-Browne NB. 2012. Memory-guidedattention: control from multiple memory systems. TrendsCogn Sci 16: 576–579.

Iaria G, Petrides M, Dagher A, Pike B, Bohbot VD. 2003.Cognitive strategies dependent on the hippocampus andcaudate nucleus in human navigation: Variability andchange with practice. J Neurosci 13: 5945–5952.

James W. 1890. Principles of psychology. Holt, New York.

Jeneson A, Squire LR. 2012. Working memory, long-termmemory, and medial temporal lobe function. Learn Mem19: 15–25.

Jeneson A, Mauldin KN, Squire LR. 2010. Intact workingmemory for relational information after medial temporallobe damage. J Neurosci 30: 13624–13629.

Jeneson A, Mauldin KN, Hopkins RO. 2011. The role of thehippocampus in retaining relational information acrossshort delays: The importance of memory load. LearnMem 18: 301–305.

Kandel ER, Dudai Y, Mayford MR. 2014. The molecular andsystems biology of memory. Cell 157: 163–186.

Kim JJ, Lee HJ, Han JS, Packard MG. 2001. Amygdala iscritical for stress-induced modulation of hippocampallong-term potentiation and learning. J Neurosci 21:5222–5228.

Kim S, Sapiurka M, Clark R, Squire LR. 2013. Contrastingeffects on path integration after hippocampal damage inhumans and rats. Proc Natl Acad Sci 110: 4732–4737.

Knowlton BJ, Mangels JA, Squire LR. 1996. A neostriatalhabit learning system in humans. Science 273: 1399–1402.

Konishi K, Etchamendy N, Roy S, Marighetto A, Rajah N,Bohbot VD. 2013. Decreased functional magnetic reso-nance imaging activity in the hippocampus in favor of thecaudate nucleus in older adults tested in a virtual navi-gation task. Hippocampus 23: 1005–1014.

Kumaran D, Wagner AD. 2009. It’s in my eyes, but it doesn’tlook that way to me. Neuron 63: 561–563.

LeDoux J. 2014. Coming to terms with fear. Proc Natl AcadSci 111: 2871–2878.

Levy DA, Stark CEL, Squire LR. 2004. Intact conceptualpriming in the absence of declarative memory PsycholSci 15: 680–685.

Liljeholm M, O’Doherty JP. 2012. Contributions of the stri-atum to learning, motivation, and performance: An as-sociative account. Trends Cogn Sci 16: 467–475.





Maccotta L, Buckner RL. 2004. Evidence for neural effects ofrepetition that directly correlate with behavioral priming.J Cog Neurosci 16: 1625–1632.

Manns JR, Squire LR. 2001. Perceptual learning, awareness,and the hippocampus. Hippocampus 11: 776–782.

Manns JM, Clark RE, Squire LR. 2001. Single-cue delayeyeblink classical conditioning is unrelated to awareness.Cogn Affect Behav Neurosci 1: 192–198.

Markowitsch H. 1988. Diencephalic amnesia: A reorienta-tion towards tracts? Brain Res Rev 13: 351–370.

McDonald RJ, Hong NS. 2013. How does a specific learningand memory system in the mammalian brain gain con-trol of behavior? Hippocampus, 23: 1084–1102.

McDougall W. 1923. Outline of psychology. Scribners,New York.

McGaugh JL, Roozendaal B. 2009. Drug enhancement ofmemory consolidation: Historical perspective and neu-robiological implications. Psychopharmacology 1: 3–14.

McMahon DB, Olson CR. 2007. Repetition suppression inmonkey inferotemporal cortex: Relation to behavioralpriming. J Neurophysiol 97: 3532–3543.

Milad MR, Quirk GJ. 2012. Fear extinction as a model fortranslational neuroscience: Ten years of progress. Ann RevPsychol 63: 129–151.

Milner B. 1962. Les troubles de la memoire accompagnantdes lesions hippocampiques bilaterales. In Physiologie del’hippocampe, pp. 257–272. Centre National de la Re-cherche Scientifique, Paris.

Milner B. 1972. Disorders of learning and memory after tem-poral lobe lesions in man. Clin Neurosurg 19: 421–466.

Mishkin M. 1978. Memory in monkeys severely impaired bycombined but not by separate removal of amygdala andhippocampus. Nature 273: 297–298.

Mishkin M, Malamut B, Bachevalier J. 1984. Memories andhabits: Two neural systems. In Neurobiology of learningand memory (ed. Lynch G, et al.), pp. 65–77. Guilford,New York.

Moscarello JM, LeDoux JE. 2013. Active avoidance learningrequires prefrontal suppression of amygdala-mediateddefensive reactions. J Neurosci 33: 3815–3823.

Nichols EA, Kao YC, Verfaellie M, Gabrieli JD. 2006. Work-ing memory and long-term memory for faces: Evidencefrom fMRI and global amnesia for involvement of themedial temporal lobes. Hippocampus 16: 604–616.

O’Keefe J, Nadel L. 1978. The hippocampus as a cognitivemap. Oxford University Press, London.

Olson IR, Page K, Moore KS, Chatterjee A, Verfaellie M.2006. Working memory for conjunctions relies on themedial temporal lobe. J Neurosci 26: 4596–4601.

Packard MG, Goodman J. 2013. Factors that influence therelative use of multiple memory systems. Hippocampus23: 1044–1052.

Packard MG, McGaugh JL. 1996. Inactivation of hippocam-pus or caudate nucleus with lidocaine differentially af-fects expression of place and response learning. NeurobiolLearn Mem 65: 65–72.

Packard MG, Hirsh R, White NM. 1989. Differential effectsof fornix and caudate nucleus lesions on two radial mazetasks: Evidence for multiple memory systems. J Neurosci9: 1465–1472.

Paller KA, Hutson CA, Miller BB, Boehm SG. 2003. Neuralmanifestations of memory with and without awareness.Neuron 38: 507–516.

Piekema C, Kessels RP, Mars RB, et al. 2006. The right hip-pocampus participates in short-term memory mainte-nance of object–location associations. NeuroImage 33:374–382.

Poldrack RA, Gabrieli JD. 2001. Characterizing the neuralmechanisms of skill learning and repetition priming:Evidence from mirror reading. Brain 124: 67–82.

Poldrack RA, Packard MG. 2003. Competition among mul-tiple memory systems: Converging evidence from animaland human brain studies. Neuropsychologia 41: 245–251.

Poldrack RA, Rodriguez P. 2003. Sequence learning: What’sthe hippocampus to do? Neuron 37: 891–893.

Poldrack RA, Clark J, Pare-Blagoev EJ, Shohamy D, CresoMoyano J, Myers C, Gluck MA. 2001. Interactive memorysystems in the human brain. Nature 414: 546–550.

Postle BR. 2006. Working memory as an emergent propertyof the mind and brain. Neurosci 139: 23–38.

Ranganath C, Blumenfeld RS. 2005. Doubts about doubledissociations between short- and long-term memory.Trends Cogn Sci 9: 374–380.

Ranganath C, D’Esposito M. 2001. Medial temporal lobeactivity associated with active maintenance of novel in-formation. Neuron 31: 865–873.

Ranganath C, Cohen MX, Brozinsky CJ. 2005. Workingmemory maintenance contributes to long-term memoryformation: Neural and behavioral evidence. J Cogn Neu-rosci 17: 994–1010.

Reber PJ, Squire LR. 1994. Parallel brain systems for learningwith and without awareness. Learn Mem 1: 217–229.

Rissman J, Gazzaley A, D’Esposito M. 2008. Dynamic ad-justments in prefrontal, hippocampal, and inferior tem-poral interactions with increasing visual working mem-ory load. Cerebr Cortex 18: 1618–1629.

Rose M, Haider H, Weiller C, Buchel C. 2002. The role ofmedial temporal lobe structures in implicit learning: Anevent-related fMRI study. Neuron 36: 1221–1231.

Ryle G. 1949. The concept of mind. Hutchinson, San Fran-cisco.

Schacter DL, Buckner RL. 1998. Priming and the brain.Neuron 20: 185–195.

Schacter DL, Wig GS, Stevens WD. 2007. Reductions incortical activity during priming. Curr Opin Neurobiol17: 171–176.

Schendan HE, Searl MM, Melrose RJ, Stern CE. 2003. AnfMRI study of the role of the medial temporal lobe inimplicit and explicit sequence learning. Neuron 37:1013–1025.

Schon D, Lorber B, Spacal M, Semenza C. 2004. A selectivedeficit in the production of exact musical intervals fol-lowing right-hemisphere damage. Cogn Neuropsychol 21:773–784.

Schultz W. 2013. Updating dopamine reward signals. CurrOpin Neurobiol 23: 229–238.

Schwabe L. 2013. Stress and the engagement of multiplememory systems: Integration of animal and human stud-ies. Hippocampus 23: 1035–1043.





Scoville WB, Milner B. 1957. Loss of recent memory afterbilateral hippocampal lesions. J Neurol Neurosurg Psychi-atry 20: 11–21.

Sherry DF, Schacter DL. 1987. The evolution of multiplememory systems. Psychol Rev 94: 439–454.

Shimamura AP, Janowsky JS, Squire LR. 1991. What is therole of frontal lobe damage in memory disorders? InFrontal lobe functioning and dysfunction (ed. Levin HD,et al.), pp. 173–195. Oxford University Press, New York.

Shohamy D, Turk-Browne NB. 2013. Mechanisms for wide-spread hippocampal involvement in cognition. J Exp Psy-chol Gen 142: 1159–1170.

Shrager Y, Levy DA, Hopkins RO, Squire LR. 2008. Workingmemory and the organization of brain systems. J Neurosci28: 4818–4822.

Smith KS, Graybiel AM. 2013. A dual operator view of ha-bitual behavior reflecting cortical and striatal dynamics.Neuron 79: 361–374.

Smith C, Squire LR. 2005. Declarative memory, awareness,and transitive inference. J Neurosci 25: 10138–10146.

Smith CN, Squire LR. 2008. Experience-dependent eyemovements reflect hippocampus-dependent (aware)memory. J Neurosci 28: 12825–12833.

Squire LR. 2009. Memory and brain systems: 1969–2009. JNeurosci 29: 12711–12716.

Squire LR, Wixted JT. 2011. The cognitive neuroscience ofhuman memory since H.M. Ann Rev Neurosci 34: 259–288.

Squire LR, Zola-Morgan S. 1988. Memory: Brain systemsand behavior. Trends Neurosci 11: 170–175.

Squire LR, Zola-Morgan S. 1991. The medial temporal lobememory system. Science 253: 1380–1386.

Squire LR, Ojemann JG, Miezin FM, Petersen SE, VideenTO, Raichle ME. 1992. Activation of the hippocampus innormal humans: A functional anatomical study of mem-ory. Proc Natl Acad Sci 89: 1837–1841.

Squire LR, Stark CEL, Clark RE. 2004. The medial temporallobe. Ann Rev Neurosci 27: 279–306.

Staresina BP, Duncan KD, Davachi L. 2011. Perirhinal andparahippocampal cortices differentially contribute to lat-er recollection of object- and scene-related event details. JNeurosci 31: 8739–8747.

Theeuwes J. 2013. Feature-based attention: It is all bottom-up priming. Phil Trans R Soc B Biol Sci 368: 20130055.

Thompson RF, Krupa DJ. 1994. Organization of memorytraces in the mammalian brain. Ann Rev Neurosci 17:519–550.

Thompson RF, Steinmetz JE. 2009. The role of the cerebel-lum in classical conditioning of discrete behavioral re-sponses. Neurosci 162: 732–755.

Thorn CA, Atallah H, Howe H, Graybiel AM. 2010. Differ-ential dynamics of activity changes in dorsolateral anddorsomedial striatal loops during learning. Neuron 66:781–795.

Toepper M, Markowitsch HJ, Gebhardt H, Beblo T, ThomasC, Gallhofer B, Driessen M, Sammer G. 2010. Hippocam-pal involvement in working memory encoding of chang-ing locations: An fMRI study. Brain Res 1354: 91–99.

Tolman RC. 1948. Cognitive maps in rats and man. PsycholRev 55: 189–208.

Tulving E. 1983. Elements of episodic memory. Oxford Uni-versity Press, Cambridge, MA.

Tulving E. 1989. Remembering and knowing the past. Amer-ican Scientist 77: 361–367.

Tulving E. 2005. Episodic memory and autonoesis: Unique-ly human? In The missing link in cognition: Evolution ofself-knowing consciousness (ed. Terrace H, et al.), pp. 3–56. Oxford University Press, New York.

Tulving E, Schacter DL. 1990. Priming and human memorysystems. Science 247: 301–306.

Victor M, Adams RD, Collins GH. 1989. The Wernicke–Korsakoff syndrome and related neurological disordersdue to alcoholism and malnutrition. Davis, Philadelphia.

Warren DE, Duff MC, Tranel D, Cohen NJ. 2011. Observingdegradation of visual representations over short intervalswhen medial temporal lobe is damaged. J Cogn Neurosci23: 3862–3873.

Warren DE, Duff MC, Jensen U, Tranel D, Cohen NJ. 2012.Hiding in plain view: Lesions of the medial temporal lobeimpair online representation. Hippocampus 22: 1577–1588.

Westerberg CE, Miller BB, Reber PJ, Cohen NJ, Paller KA.2011. Neural correlates of contextual cueing are modulat-ed by explicit learning. Neuropsychologia 49: 3439–3447.

Whitlock JR, Sutherland RJ, Witter MP, Moser MB, MoserEI. 2008. Navigating from hippocampus to parietal cor-tex. Proc Natl Acad Sci 105: 14755–14762.

Wiggs CL, Martin A. 1998. Properties and mechanisms ofperceptual priming. Curr Opin Neurobiol 8: 227–233.

Wilkins LK, Girard TA, Konishi K, King M, Herdman KA,King J, Christensen B, Bohbot VD. 2013. Selective deficitin spatial memory strategies contrast to intact responsestrategies in patients with schizophrenia spectrum disor-ders tested in a virtual navigation task. Hippocampus23: 1015–1024.

Winograd T. 1975. Frame representations and the declara-tive-procedural controversy. In Representation and under-standing: Studies in cognitive science (ed. Bobrow D, etal.), pp. 185–210. Academic, New York.

Yee LT, Hannula DE, Tranel D, Cohen NJ. 2014. Short-termretention of relational memory in amnesia revisited: Ac-curate performance depends on hippocampal integrity.Front Hum Neurosci 8: 16.

Yin HH, Knowlton BJ. 2006. The role of the basal ganglia inhabit formation. Nat Rev Neurosci 7: 464–476.

Yin HH, Mulcare SP, Hilario MRF, Clouse E, Holloway T,Davis MI, Hansson AC, Lovinger DM, Costa RM, et al.2009. Dynamic reorganization of striatal circuits duringthe acquisition and consolidation of a skill. Nat Neurosci12: 333–341.

Zaki SR. 2004. Is categorization performance really intact inamnesia? A meta-analysis. Pysch Bull Rev 11: 1048–1054.

Zaki SR, Nosofsky RM, Jessup NM, Unversagt FR. 2003.Categorization and recognition performance of a mem-ory-impaired group: Evidence for single system models. JInt Neuropsychol Soc 9: 94–406.





2015; doi: 10.1101/cshperspect.a021667Cold Spring Harb Perspect Biol Larry R. Squire and Adam J.O. Dede Conscious and Unconscious Memory Systems

Subject Collection Learning and Memory

Conscious and Unconscious Memory SystemsLarry R. Squire and Adam J.O. Dede

Nonassociative Learning in InvertebratesJohn H. Byrne and Robert D. Hawkins

Place Cells, Grid Cells, and Memory

MoserMay-Britt Moser, David C. Rowland and Edvard I. Consolidation

The Regulation of Transcription in Memory

Cristina M. Alberini and Eric R. Kandel

http://cshperspectives.cshlp.org/cgi/collection/ For additional articles in this collection, see

Copyright © 2015 Cold Spring Harbor Laboratory Press; all rights reserved


http://cshperspectives.cshlp.org/cgi/collection/

http://cshperspectives.cshlp.org/cgi/collection/

http://cshperspectives.cshlp.org/cgi/adclick/?ad=42791&adclick=true&url=http%3A%2F%2Fwww.genelink.com%2Fnewsite%2Fproducts%2Frnaoligos.asp


conscious and unconscious memory...

Documents